Find in Library

Islet autoantibody (iAb)-positive individuals have a high risk of progression to type 1 diabetes (T1D), although the rate of progression is highly variable and factors involved in the rate of progression are largely unknown. The ratio of unmethylated/methylated insulin DNA levels (unmethylated <i>INS</i> ratio) has been shown to be higher in participants at high risk of T1D compared to healthy controls. We aimed to evaluate whether an unmethylated <i>INS</i> ratio may be a useful biomarker of beta cell death and rate of progression to T1D. In TrialNet participants who were followed in the Pathway to Prevention Study and progressed to diabetes (<i>n</i> = 57, median age of onset 15.3 years), we measured unmethylated <i>INS</i> ratio and autoantibodies by electrochemiluminescence (ECL) assays (ECL-IAA, ECL-GADA, and ECL-IA2) and radioimmunoassays (RIA) (mIAA, GADA, IA2A, and ZnT8A) longitudinally for 24 months prior to diagnosis. Linear models were used to test the association between unmethylated <i>INS</i> ratio and the age at T1D diagnosis and unmethylated <i>INS</i> ratio and iAb over time. Close to diabetes onset, the unmethylated <i>INS</i> ratio was associated with mIAA (<i>p</i> = 0.003), ECL-IAA (<i>p</i> = 0.002), and IA2A (<i>p</i> = 0.01) levels, but not with GADA, ECL-GADA, ECL-IA2, or ZnT8A levels. No significant associations were found at baseline (24 months prior to T1D diagnosis). Only mIAA levels were significantly associated with an unmethylated <i>INS</i> ratio over time, with a 0.24 change in the ratio for each 0.1 change in mIAA z-score (<i>p</i> = 0.02). Adjusting for a baseline unmethylated <i>INS</i> ratio, an increased rate of change in unmethylated <i>INS</i> ratio from baseline to diabetes onset was associated with a five-year decrease in age at T1D diagnosis (<i>p</i> = 0.04).